From: allenai <github.com/allenai>Newsgroups: alt.freeweights.toolsSubject:

olmOCR

Release: v0.4.27 (2026-03-12)Commit: 1e139a5License: Apache-2.0

Converts PDFs to clean text with a vision-language model, built for training-data scale.

data and documentsv0.4.27releaseolmOCRPythonApache-2.0
FactsLicenseApache-2.0LanguagePythonKinddata and documentsRelease2026-03-12PlatformsLinux (GPU)Links to the projectpermissive
$ pip install olmocr[gpu]

Overview

Converts PDFs to clean text with a vision-language model, built for training-data scale.

Kind
data and documents
License
Apache-2.0
Language
Python
Platforms
Linux (GPU)
Latest release
v0.4.27 (2026-03-12)
Repository
allenai/olmocr

Freeweights links to the project and does not host tool binaries.

Install

$ pip install olmocr[gpu]
$ pip install olmocr[gpu]

Checked against the project's README on 2026-11-14. The project's own instructions win when they differ: install docs

Freeweights links to the project and does not host tool binaries.

Use with Freeweights

Use a catalog model

Download the weights from Freeweights first (HF_ENDPOINT or fw.py), then point the tool at the folder.

Catalog models it can run (1206)

and 1146 more.

License

License
Apache-2.0
Class
Permissive

Read the license

Freeweights links to the project and does not host tool binaries.

Provenance

From: allenai <github.com/allenai>Newsgroups: alt.freeweights.toolsSubject: olmOCRRelease: v0.4.27 (2026-03-12)Commit: 1e139a5License: Apache-2.0
Repository
allenai/olmocr
Pinned release
v0.4.27, 2026-03-12
Release commit
1e139a5ea61f1668164e6b63357d7284a8391615
Default branch
main f7cfe4c22098b154c76b6ec950d1c0a464eecf8d
Fetched
2026-11-14
Takedowns
None.

Freeweights links to the project and does not host tool binaries.

Works with

1206 catalog models. See the list